Project Summary Understanding the circuit mechanisms that give rise to perception and behavior requires linking neuronal activity to connectivity. This can be accomplished at multiple scales and ideally can be related to further studies using activity manipulations to demonstrate causality. Recent work in the mouse visual system has revealed the contributions of specific cell types to the generation of visual receptive field properties as well as state-dependent changes in the representation of visual information. But it is unknown whether the cortical circuit mechanisms and principles revealed in the mouse are preserved across species. This project aims to develop and refine molecular, genetic, viral and large scale optical and electrical recording tools for use in the non-rodent cortex. Paradigms will be established by which these tools can link visual function to cortical modules, cell types, and connectivity. Experiments that expand knowledge of the role of specific cortical cell types to be comparable to data collected in mice are required to evaluate what are common circuit mechanisms and principles of cell type specific computations versus circuits that are specialized to particular species or functions. Systematically controlling visual stimuli while conducting recordings of activity with these tools will create data sets that make it possible to test whether principles and functions of specific circuit motifs emerging from studies in the mouse cortex can be generalized to higher species. Specific aims are organized around levels of selectivity at which visually-evoked activity will be linked to circuits: 1) modules, 2) cell types, and 3) connectivity. These aims share two different basic approaches for recording dynamic activity from large neuronal populations ? two-photon calcium imaging and high-density (128 and 384 channel) laminar silicone electrode arrays. Aim 1 will link visually evoked neuronal activity to modular and laminar organization of primary visual cortex (V1). This knowledge can be combined with known relationships between connectivity and modular/laminar organization to link circuits to function. Aim 2 will link visually evoked neuronal activity to V1 cell types by: combining 2-photon calcium imaging with post mortem identification and antibody staining; and recording activity of single neurons with high-density laminar electrode arrays and then identifying cell types based on electrical images. Aim 3 will directly link visually evoked neuronal activity to connectivity using cross-correlation analysis of recordings from up to 150 neurons recorded simultaneously with high-density laminar electrode arrays.